Re: [Lse-tech] [RFC] sched_clock() and "drifty" timebases

[john s. <joh...@us...>]

On Thu, 2003-11-20 at 14:48, John Hawkes wrote:
> I proposed a patch to implement a platform-specific ia64 sched_clock() for use
> by SGI's SN platform, and David Mosberger countered with a desire for a more
> general solution.  He suggests something along the lines of separating the
> current use of sched_clock() into potentially two different timestamps for
> "drifty timebase" platforms:
> 1) Continuing to use a high-resolution timestamp, using a local timebase, for
> purpose #1.
> 2) For drifty platforms, separately using a low-resolution "jiffies"-based
> timestamp, for purpose #2.  Architectures and platforms without drift would
> continue to use the current sched_clock() for both #1 and #2.  Drifty
> platforms would (by use of macros) implement separate timestamps.
> 
> This new scheme would allow i386-NUMA to use a high-resolution timebase for
> purpose #1.  Do the i386-NUMA folks have any interest in this?

Hmmm. I'm worried about having a single function that does different
things in different settings. Why not use get_cycles() when we want to
read the cycle count, and use sched_clock when we want some sort of "as
high a resolution that's safe" abstraction?

Regardless, right now I'm feeling very conservative, and I'm not sure
how much the added complexity would win. So for 2.6.0 I think we're ok
using jiffies, but I'd be interested in playing with implementations of
your proposed scheme. 


thanks
-john